Technical report on assessment of rumble inversion results (DE18)

The main goal of project RUMBLE has been to enhance our capability to predict active sonar detection ranges in shallow water. The idea is to measure ‘through-the-sensor’ the seabed characteristics that affects long-range acoustic propagation in shallow water. The work has included the development of an inversion method for determining bottom parameters from reverberation received by a Low frequency Active Sonar (LFAS). Two sea trials have also been performed, in order to provide data for evaluating the performance of the method. The first sea trial took place in a relatively simple area in Vestfjorden, while for the second sea trial a more difficult area in the Norwegian Trench, west of Stavanger, was chosen. The purpose of the present report is to make an assessment of the inversion results; to validate the results against ground truth, and to estimate the expected improvement of the RUMBLE method over predictions using standard databases. It was found that reliable measurements could be made of the changes in backscattering strength, but not of sound speed or reflection loss. Ground truth collected during the trial, grab samples and 38 kHz echosounder, were found to be only weakly correlated with the LFAS measurements. This shows that near surface properties are of little use as an indicator of the effect of the seabed on LFAS.